The present disclosure relates to Internet of Things devices such as smart appliances, and more specifically, to normalizing heterogeneous information in labels processed by the IoT devices.
Internet of Things (IoT) is an approach that networks physical objects that are embedded with electronics, software, sensors, and the like that enables these objects to collect and exchange data. IoT devices are typically networked via a hub that allows messages to be sent back and forth between the hub and devices. Further, cloud computing platforms can provide multiple services for networked IoT devices, such as IoT service applications and cloud storage of collected data.
Common household appliances may implement IoT technology to improve a user experience for an individual. As an example, an IoT-based washing machine may collect data relating to clothes being loaded into the washing machine. To do so, the washing machine may be configured with sensors, such as a camera and an RFID reader, that scan labels provided with the clothing. As a result, the washing machine can automatically identify information associated with a given clothing item inserted into the washing machine, such as a clothing type, clothing color, care instructions, and the like. Based on the information, the washing machine can then automatically determine settings to perform relative to the clothing item, such as a minimum or maximum temperature that can be set, an amount of fabric softener to dispense, and the like.
One concern with appliances that scan such data from input items (like clothes in a laundry machine) is that a label accompanying a given item can differ significantly from a label of another similar item. For example, a shirt A can have a clothing label that is printed in English with instructions that use British units (e.g., Fahrenheit temperatures), whereas a shirt B can have a clothing label that is printed in Chinese with instructions that use metric units (e.g., centigrade temperatures). Further, the labels themselves may be provided with the clothing using different methods, such as through storing information in RFID tags, printing label information on the fabric directly, providing label information as a URL printed on the fabric, etc. Based on the variety of labels used for the items, the appliance might be unable to identify how to process a given item consistently.